In relentless pursuit of the white whale: A role for the ventral midline thalamus in behavioral flexibility and adaption?

The reuniens (Re) nucleus is located in the ventral midline thalamus. It has fostered increasing interest, not only for its participation in a variety of cognitive functions (e.g., spatial working memory, systemic consolidation, reconsolidation, extinction of fear or generalization), but also for its neuroanatomical positioning as a bidirectional relay between the prefrontal cortex (PFC) and the hippocampus (HIP). In this review we compile and discuss recent studies having tackled a possible implication of the Re nucleus in behavioral flexibility, a major PFC-dependent executive function controlling goal-directed behaviors. Experiments considered explored a possible role for the Re nucleus in perseveration, reversal learning, fear extinction, and set-shifting. They point to a contribution of this nucleus to behavioral flexibility, mainly by its connections with the PFC, but possibly also by those with the hippocampus, and even with the amygdala, at least for fear-related behavior. As such, the Re nucleus could be a crucial crossroad supporting a PFC-orchestrated ability to cope with new, potentially unpredictable environmental contingencies, and thus behavioral flexibility and adaption.

Disconnecting prefrontal cortical neurons from the ventral midline thalamus: Loss of specificity due to progressive neural toxicity of an AAV-Cre in the rat thalamus.

BACKGROUND: The thalamic reuniens (Re) and rhomboid (Rh) nuclei are bidirectionally connected with the medial prefrontal cortex (mPFC) and the hippocampus (Hip). Fiber-sparing N-methyl-D-aspartate lesions of the ReRh disrupt cognitive functions, including persistence of certain memories. Because such lesions irremediably damage neurons interconnecting the ReRh with the mPFC and the Hip, it is impossible to know if one or both pathways contribute to memory persistence. Addressing such an issue requires selective, pathway-restricted and direction-specific disconnections. NEW METHOD: A recent method associates a retrograde adeno-associated virus (AAV) expressing Cre recombinase with an anterograde AAV expressing a Cre-dependent caspase, making such disconnection feasible by caspase-triggered apoptosis when both constructs meet intracellularly. We injected an AAVrg-Cre-GFP into the ReRh and an AAV5-taCasp into the mPFC. As expected, part of mPFC neurons died, but massive neurotoxicity of the AAVrg-Cre-GFP was found in ReRh, contrasting with normal density of DAPI staining. Other stainings demonstrated increasing density of reactive astrocytes and microglia in the neurodegeneration site. COMPARISON WITH EXISTING METHODS: Reducing the viral titer (by a 4-fold dilution) and injection volume (to half) attenuated toxicity substantially, still with evidence for partial disconnection between mPFC and ReRh. CONCLUSIONS: There is an imperative need to verify potential collateral damage inherent in this type of approach, which is likely to distort interpretation of experimental data. Therefore, controls allowing to distinguish collateral phenotypic effects from those linked to the desired disconnection is essential. It is also crucial to know for how long neurons expressing the Cre-GFP protein remain operational post-infection.

Inhibition of the ventral midline thalamus does not alter encoding, short-term holding or retrieval of spatial information in rats performing a water-escape working memory task.

Working memory (WM) is a function operating in three successive phases: encoding (sample trial), holding (delay), and retrieval (test trial) of information. Studies point to a possible implication of the thalamic reuniens nucleus (Re) in spatial WM (SWM). In which of the aforementioned 3 phases the Re has a function is largely unknown. Recently, in a delayed SWM water-escape task, we found that performance during the retrieval trial correlated positively with c-Fos expression in the Re nucleus, suggesting participation in retrieval. Here, we used the same task and muscimol (MUSC) inhibition or DREADD(hM4Di)-mediated inhibition of the Re during information encoding, right thereafter (thereby affecting the holding phase), or during the retrieval trial. A 6-hour delay separated encoding from retrieval. Concerning SWM, MUSC in the Re nucleus did not alter performance, be it during or after encoding, or during evaluation. CNO administered before encoding in DREADD-expressing rats was also ineffective, although CNO-induced inhibition disrupted set shifting performance, as found previously (Quet et al., Brain Struct Function 225, 2020), thereby validating DREADD efficiency. These findings are the first that do not support an implication of the Re nucleus in SWM. As most previous studies used T-maze alternation tasks, which carry high proactive interference risks, an important question to resolve now is whether the Re nucleus is required in (T-maze alternation) tasks using very short information-holding delays (seconds to minutes), and less so in other short-term spatial memory tasks with longer information holding intervals (hours) and therefore reduced interference risks.

Ventral midline thalamus activation is correlated with memory performance in a delayed spatial matching-to-sample task: A c-Fos imaging approach in the rat.

The reuniens (Re) and rhomboid (Rh) nuclei of the ventral midline thalamus are bi-directionally connected with the hippocampus and the medial prefrontal cortex. They participate in a variety of cognitive functions, including information holding for seconds to minutes in working memory tasks. What about longer delays? To address this question, we used a spatial working memory task in which rats had to reach a platform submerged in water. The platform location was changed every 2-trial session and rats had to use allothetic cues to find it. Control rats received training in a typical response-memory task. We interposed a 6 h interval between instruction (locate platform) and evaluation (return to platform) trials in both tasks. After the last session, rats were killed for c-Fos imaging. A home-cage group was used as additional control of baseline levels of c-Fos expression. C-Fos expression was increased to comparable levels in the Re (not Rh) of both spatial memory and response-memory rats as compared to their home cage counterparts. However, in spatial memory rats, not in their response-memory controls, task performance was correlated with c-Fos expression in the Re: the higher this expression, the better the performance. Furthermore, we noticed an activation of hippocampal region CA1 and of the anteroventral nucleus of the rostral thalamus. This activation was specific to spatial memory. The data point to a possible performance-determinant participation of the Re nucleus in the delayed engagement of spatial information encoded in a temporary memory.

The reuniens and rhomboid nuclei of the thalamus: A crossroads for cognition-relevant information processing?

Over the past twenty years, the reuniens and rhomboid (ReRh) nuclei, which constitute the ventral midline thalamus, have received constantly growing attention. Since our first review article about the functional contributions of ReRh nuclei (Cassel et al., 2013), numerous (>80) important papers have extended anatomical knowledge, including at a developmental level, introduced new and very original electrophysiological insights on ReRh functions, and brought novel results on cognitive and non-cognitive implications of the ReRh. The current review will cover these recent articles, more on Re than on Rh, and their contribution will be approached according to their affiliation with work before 2013. These neuroanatomical, electrophysiological or behavioral findings appear coherent and point to the ReRh nuclei as two major components of a multistructural system supporting numerous cognitive (and non-cognitive) functions. They gate the flow of information, perhaps especially from the medial prefrontal cortex to the hippocampus and back, and coordinate activity and processing across these two (and possibly other) brain regions of major cognitive relevance.

The nucleus reuniens, a thalamic relay for cortico-hippocampal interaction in recent and remote memory consolidation.

The consolidation of declarative memories is believed to occur mostly during sleep and involves a dialogue between two brain regions, the hippocampus and the medial prefrontal cortex. The information encoded during experience by neuronal assemblies is replayed during sleep leading to the progressive strengthening and integration of the memory trace in the prefrontal cortex. The gradual transfer of information from the hippocampus to the medial prefrontal cortex for long-term storage requires the synchronization of cortico-hippocampal networks by different oscillations, like ripples, spindles, and slow oscillations. Recent studies suggest the involvement of a third partner, the nucleus reuniens, in memory consolidation. Its bidirectional connections with the hippocampus and medial prefrontal cortex place the reuniens in a key position to relay information between the two structures. Indeed, many topical works reveal the original role that the nucleus reuniens occupies in different recent and remote memories consolidation. This review aimed to examine these contributions, as well as its functional embedment in this complex memory network, and provide some insights on the possible mechanisms.

Ventral midline thalamus is not necessary for systemic consolidation of a social memory in the rat.

According to the standard theory of memory consolidation, recent memories are stored in the hippocampus before their transfer to cortical modules, a process called systemic consolidation. The ventral midline thalamus (reuniens and rhomboid nuclei, ReRh) takes part in this transfer as its lesion disrupts systemic consolidation of spatial and contextual fear memories. Here, we wondered whether ReRh lesions would also affect the systemic consolidation of another type of memory, namely an olfaction-based social memory. To address this question we focused on social transmission of food preference. Adult Long-Evans rats were subjected to N-methyl-d-aspartate-induced, fibre-sparing lesions of the ReRh nuclei or to a sham-operation, and subsequently trained in a social transmission of food preference paradigm. Retrieval was tested on the next day (recent memory, nSham = 10, nReRh = 12) or after a 25-day delay (remote memory, nSham = 10, nReRh = 10). All rats, whether sham-operated or subjected to ReRh lesions, learned and remembered the task normally, whatever the delay. Compared to our former results on spatial and contextual fear memories (Ali et al., 2017; Klein et al., 2019; Loureiro et al., 2012; Quet et al., 2020), the present findings indicate that the ReRh nuclei might not be part of a generic, systemic consolidation mechanism processing all kinds of memories in order to make them persistent. The difference between social transmission of food preference and spatial or contextual fear memories could be explained by the fact that social transmission of food preference is not hippocampus-dependent and that the persistence of social transmission of food preference memory relies on different circuits.

The reuniens and rhomboid nuclei are necessary for contextual fear memory persistence in rats.

Memory persistence refers to the process by which a temporary, labile memory is transformed into a stable and long-lasting state. This process involves a reorganization of brain networks at systems level, which requires functional interactions between the hippocampus (HP) and medial prefrontal cortex (mPFC). The reuniens (Re) and rhomboid (Rh) nuclei of the ventral midline thalamus are bidirectionally connected with both regions, and we previously demonstrated their crucial role in spatial memory persistence. We now investigated, in male rats, whether specific manipulations of ReRh activity also affected contextual and cued fear memory persistence. We showed that the permanent ReRh lesion impaired remote, but not recent contextual fear memory. Tone-cued recent and remote fear memory were spared by the lesion. In intact rats, acute chemogenetic ReRh inhibition conducted before recall of either recent or remote contextual fear memories produced no effect, indicating that the ReRh nuclei are not required for retrieval of such memories. This was also suggested by a functional cellular imaging approach, as retrieval did not alter c-fos expression in the ReRh. Collectively, these data are compatible with a role for the ReRh in 'off-line' consolidation of a contextual fear memory and support the crucial importance of ventral midline thalamic nuclei in systems consolidation of memories.

Delta opioid receptors expressed in forebrain GABAergic neurons are responsible for SNC80-induced seizures.

The delta opioid receptor (DOR) has raised much interest for the development of new therapeutic drugs, particularly to treat patients suffering from mood disorders and chronic pain. Unfortunately, the prototypal DOR agonist SNC80 induces mild epileptic seizures in rodents. Although recently developed agonists do not seem to show convulsant properties, mechanisms and neuronal circuits that support DOR-mediated epileptic seizures remain to be clarified. DORs are expressed throughout the nervous system. In this study we tested the hypothesis that SNC80-evoked seizures stem from DOR activity at the level of forebrain GABAergic transmission, whose inhibition is known to facilitate the development of epileptic seizures. We generated a conditional DOR knockout mouse line, targeting the receptor gene specifically in GABAergic neurons of the forebrain (Dlx-DOR). We measured effects of SNC80 (4.5, 9, 13.5 and 32 mg/kg), ARM390 (10, 30 and 60 mg/kg) or ADL5859 (30, 100 and 300 mg/kg) administration on electroencephalograms (EEGs) recorded in Dlx-DOR mice and their control littermates (Ctrl mice). SNC80 produced dose-dependent seizure events in Ctrl mice, but these effects were not detected in Dlx-DOR mice. As expected, ARM390 and ADL5859 did not trigger any detectable change in mice from both genotypes. These results demonstrate for the first time that SNC80-induced DOR activation induces epileptic seizures via direct inhibition of GABAergic forebrain neurons, and supports the notion of differential activities between first and second-generation DOR agonists.

Clusters of cerebellar Purkinje cells control their afferent climbing fiber discharge.

Climbing fibers, the projections from the inferior olive to the cerebellar cortex, carry sensorimotor error and clock signals that trigger motor learning by controlling cerebellar Purkinje cell synaptic plasticity and discharge. Purkinje cells target the deep cerebellar nuclei, which are the output of the cerebellum and include an inhibitory GABAergic projection to the inferior olive. This pathway identifies a potential closed loop in the olivo-cortico-nuclear network. Therefore, sets of Purkinje cells may phasically control their own climbing fiber afferents. Here, using in vitro and in vivo recordings, we describe a genetically modified mouse model that allows the specific optogenetic control of Purkinje cell discharge. Tetrode recordings in the cerebellar nuclei demonstrate that focal stimulations of Purkinje cells strongly inhibit spatially restricted sets of cerebellar nuclear neurons. Strikingly, such stimulations trigger delayed climbing-fiber input signals in the stimulated Purkinje cells. Therefore, our results demonstrate that Purkinje cells phasically control the discharge of their own olivary afferents and thus might participate in the regulation of cerebellar motor learning.

A new neuronal target for ß-amyloid peptide in the rat hippocampus.

In Alzheimer's disease, amyloid beta peptide (Aß) accumulation is associated with hippocampal network dysfunction. Intrahippocampal injections of Aß induce aberrant inhibitory septohippocampal (SH) network activity in vivo and impairment of memory processing. In the present study, we observed, after hippocampal Aß treatment, a selective loss of neurons projecting to the medial septum (MS) and containing calbindin (CB) and/or somatostatin (SOM). Other GABAergic neuronal subpopulations were not altered. Thus, the present study identifies hippocamposeptal neuron populations as specific targets for Aß deposits. We observed that in Aß-treated rats but not in controls, glutamate agonist application induced rhythmic bursting in 55% of the slow-firing neurons in the medial septum. This suggests that hippocampal Aß can trigger modifications of the septohippocampal pathway via the alteration of a specific neuronal population. Long-range hippocamposeptal GABA/calbindin neurons, targets of hippocampal amyloid deposits, are implicated in supporting network synchronization. By identifying this target, we contribute to the understanding of the mechanisms underlying deleterious effects of Aß, one of the main agents of dementia in Alzheimer's disease.

Decreased rhythmic GABAergic septal activity and memory-associated theta oscillations after hippocampal amyloid-beta pathology in the rat.

The memory deficits associated with Alzheimer's disease result to a great extent from hippocampal network dysfunction. The coordination of this network relies on theta (symbol) oscillations generated in the medial septum. Here, we investigated in rats the impact of hippocampal amyloid beta (Abeta) injections on the physiological and cognitive functions that depend on the septohippocampal system. Hippocampal Abeta injections progressively impaired behavioral performances, the associated hippocampal theta power, and theta frequency response in a visuospatial recognition test. These alterations were associated with a specific reduction in the firing of the identified rhythmic bursting GABAergic neurons responsible for the propagation of the theta rhythm to the hippocampus, but without loss of medial septal neurons. Such results indicate that hippocampal Abeta treatment leads to a specific functional depression of inhibitory projection neurons of the medial septum, resulting in the functional impairment of the temporal network.

Learning deficits and dysfunctional synaptic plasticity induced by aggregated amyloid deposits in the dentate gyrus are rescued by chronic treatment with indomethacin.

The amyloid pathology in Alzheimer's disease is accompanied by a chronic inflammatory response characterized by gliosis and activated microglial cells surrounding senile plaques. Epidemiological studies have shown nonsteroidal anti-inflammatory drug treatment reduces the risk of Alzheimer's disease. We have previously shown that injection of a combination of Abeta40 and Abeta43 in the dentate gyrus of the rat induces aggregated amyloid deposits and inflammation associated with dysfunctional synaptic plasticity and learning deficits. Here we characterize the effectiveness of nonsteroidal anti-inflammatory treatment in this model and show that this treatment restores the working memory deficit and decremental long-term potentiation in the dentate gyrus. Importantly, we observe no qualitative difference in the presence of aggregated material but a substantial reduction in microglial-induced inflammation, suggesting that mature aggregated plaques may not be directly responsible for the deficits but may trigger an inflammatory response which has a detrimental effect on synaptic function and memory.